Instance level image translation technology based on deep attention generative adversarial network

Abstract

The invention puts forward an instance level image translation technology based on a deep attention generative adversarial network. The main content of the technology comprises a network module, an instance level image translation, collection level image translation and an integral target function. The technology comprises the following steps that: firstly, using a deep attention coder, a generator and two descriminators to construct the deep attention generative adversarial network; then, according to a given input image, adopting a positioning function to predict the position of the attention area and calculate an attention mask; then, utilizing the generator to receive a structured expression from a latent space, and generating a translation sample; and finally, using the descriminatorsto discriminate the translated sample from a real image. The invention puts forward the instance level image translation technology based on the deep attention generative adversarial network, can besimultaneously applied to instance level and collection level constraints, solves a great quantity of practical tasks and can obtain an effect of better performance.

Description

technical field [0001] The present invention relates to the field of image translation, in particular to an instance-level image translation technology based on deep attention generative confrontation network. Background technique [0002] Image translation is an application of machine translation. Users can convert text information, color information, icon information, etc. contained in an image into any form of expression. The rapid development of mobile electronic devices makes image translation technology widely used in all aspects of people's lives. For example, this technology can be used to translate the characters of other countries into the characters of one's own country, which is conducive to the transmission of information and cultural exchanges. For another example, image translation technology can be used to translate advertisements describing the nature of products that users see in shopping malls into intuitive pictures, and it can also automatically search ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems that existing methods cannot learn instance-level correspondence, the purpose of the present invention is to provide an instance-level image translation technology based on deep attention generative adversarial networks. First, a deep attention encoder, a generator and two The discriminator builds a deep attention generative adversarial network; then according to the given input image, uses the positioning function to predict the position of the attention area and calculates the attention mask; then uses the generator to receive the structured expression from the latent space to generate translation samples; finally Use the discriminator to distinguish translated samples from real images

Images